In this paper robust feedback linearization control approach based on the gap metric analysis is proposed to control a Proton Exchange Membrane Fuel Cell (PEMFC). The oxygen excess ratio ( ) is regulated through adjustment of the air supply to avoid oxygen starvation. Furthermore regulation improves the efficiency whilst more net power will be delivered. In this paper a six order state variable PEM fuel cell is used as a plant whereas the system variations and disturbances are regarded as uncertainties to configure the perturbed plant. The gap metric analysis is gained in this paper to assess the difference between the perturbed plants and that of the nominal. Results of using the nonlinear control law reveal that the proposed feedback linearization control is robust against disturbances during the oxygen excess ratio regulation.Results verify that the measurement delays in super twisting algorithm excite un-modeled dynamics because of higher frequency in the oscillations. The proposed controller eliminates influence of un-modeled dynamic and delay of actuator and sensor. Furthermore the designed controller is found capable to attenuate the practical measurement noise effect (in terms of a stochastic uncertainty) in both of the frequency spectrum and also in the overall amplitude.